| Title | Flood depth estimation from extent and topography using splines (FDEETs) |
| |
| Author | McGrath, H ;
Harrison, A; Olthof, I |
| Source | Program, 41st Canadian Symposium on Remote Sensing/Programme, 41e Symposium canadien de télédétection; 2020 p. 62  Open Access |
| Links | Online - En ligne (complete
volume - volume complet, PDF, 17.5 MB)
|
| Image |  |
| Year | 2020 |
| Alt Series | Natural Resources Canada, Contribution Series 20200563 |
| Publisher | Canadian Remote Sensing Society |
| Meeting | 41st Canadian Symposium on Remote Sensing; July 13-16, 2020 |
| Document | serial |
| Lang. | English |
| Media | on-line; digital |
| File format | pdf |
| Subjects | hydrogeology; environmental geology; geophysics; Science and Technology; Health and Safety; floods; topography; remote sensing; satellite imagery; mapping techniques; models; Flood Depth Estimation from
Extent and Topography (FDEETs); Floodwater Depth Estimation Tool (FwDET); Methodology; synthetic aperture radar surveys (SAR); Emergency services; Geographic data; Open data; Open government |
| Program | Canada Centre for Remote Sensing Flood Mapping Guidelines |
| Released | 2020 07 10 |
| Abstract | Earth Observation data is used operationally to create maps of near real-time flood extents, primarily from synthetic aperture radar (SAR) due to its ability to detect water beneath clouds. The spatial
extent of the flood extracted from SAR is important to communicate the area inundated by floodwaters. Natural Resources Canada has a team dedicated to the extraction of the near real-time flood extent during major floods from SAR data. This data
layer is available as a WMS service on the Canadian Open Government website. The near-real time flood extent is useful for identify inundated areas, but does not provide information about depth. Flood depth is an important factor to aid in emergency
response tasks involving evacuation and safety (e.g.: directing rescue and relief resources, determination of road closures and accessibility) and is an important measure for post-event analysis of property damage. In this project a new method to
derive a flood depth layer from these extents is developed and tested in order to disseminate on the Open Government site in near real-time.
Previous research has considered a variety of models and methods to generate a flood surface using flood
extent and terrain. One of the most recent approaches, FwDET, utilizes an iterative zonal approach (Cohen et al. 2017) that has been subsequently refined to improve processing speed and applicability in coastal regions by incorporating a cost-based
approach (Cohen et al. 2019). Other techniques have included cross-sections perpendicular to the flow of the river with various spacing thresholds (Scorzini et al. 2018; FEMA 2010). The FwDET model was tested in two Canadian communities and validated
against flood depth grids generated via 1-dimensional hydraulic model. Analysis of the results found many inconsistencies in depth approximation, especially in urban areas with fragmented flood extents.
To address these issues, development and
testing of Flood Depth Estimation from Extent and Topography (FDEETs) is ongoing. FDEETs utilizes a spline based spatial interpolation to generate a flood surface from a flood extent polygon which has been simplified via Douglas Peucker algorithm and
a series of equidistant river cross-sections that run perpendicular to the flow of the river. Preliminary results in Gatineau, QC and Fredericton, NB found an absolute mean water depth difference of 0.15m and 0.37m respectively across the study
areas, with the largest outliers identified at the edges of the study area. These values represent an improvement of 0.17m (Gatineau) and 0.06m (Fredericton) in the mean water difference when compared to the FwDET v2.0 solution. |
| Summary | (Plain Language Summary, not published)
Development of a technique to create flood depth layer using flood extent and digital terrain model. This data layer can be used for informing emergency
response activities and initial estimates of flood damage (eg: property damage), and could be a value-added data layer published by EGS along with the active flood layer. |
| GEOSCAN ID | 327793 |
|
|